DEFEAT ER S OF CULPA BILI T Y

145

to imposing a risk on V that will avert the threat, and if so, what are its chances of success, and how risky or difficult is it? (Whatever is the correct position on proportional response, retreat, and necessity with respect to defense against CAs, surely D has a duty to undergo some degree of sacrifice to avoid harming innocent aggressors, shields, swords, and bystanders. The excuse or personal justification of duress is premised on D’s facing a sufficiently hard choice. If D invites the hard choice by not availing herself of less difficult opportunities to avoid the threat, D herself becomes a CA against the innocent aggressors, et al.)68 How D answers these questions affects her culpability if she chooses to impose the risk on V.

4. Implications

a. Formulation: Standard versus Rules: The first issue regarding the proposed excuse concerns its formulation. We have essentially taken a “person of reasonable firmness” standard from the Model Penal Code’s defense of duress and made it the standard governing a much broader

68 See, e.g., Benbaji, supra note 53, at 598–599 n.20.

Outside of criminal gangs, it is not clear what the law envisions as culpably placing oneself in a position where duress is likely. But one possible scenario is intriguing. Suppose there is a particularly rough section of town – or some rough establishment, say a pool hall – and outsiders who venture in are frequently threatened with violence unless they commit crimes as directed by the threateners. Suppose further that an actor ventures into this area aware of the risk and is indeed threatened with death unless he mugs someone, which he does. Would the actor be one whose ability to raise duress has been lost or diminished?

Despite the law’s reference to recklessly or negligently placing oneself in situations of duress, the average reader is likely to balk at attaching this consequence to merely going where one has a legal right to go. Even if visiting a particular pool hall carries a high risk of being threatened, and not visiting it represents only a very minor setback of interests, wrongdoers should not be able to limit liberty in this way, or so one might believe.

Notice, however, that reading the law of duress to preclude the defense in such a situation is quite consistent with the requirement of retreat in the law of self-defense. That requirement entails that I must give up my liberty, on pain of forfeiting the right to employ deadly force to defend myself, rather than remain where I have a right to be. By logical extension it entails that I may not go where I otherwise have a right to go, again on pain of forfeiting the right to use deadly force in self-defense, if I am aware that in so going I am likely to be attacked. Thus, if I must retreat from the 7-Eleven to my house to avoid a deadly attack by a knife-wielding lunatic, even if in doing so I must forgo getting a Mars Bar, then likewise I should not be able to venture from my house to the store to get the Mars Bar, prepared to shoot my .45 to protect myself, if I am aware that the lunatic is there. If that is what the retreat doctrine requires regarding self-defense, the same requirement should hold for duress.

146	T HE CULPA BLE CHOICE

domain of self-protection. And the standard is quite abstract and informal, giving very uncertain guidance both to actors before the fact and to judges and juries after the fact. Why have we not proposed determinate rules instead?

If the cases of self-protection we are interested in are truly cases of excuse, then a standard rather than guidance-providing rules is appropriate. We are not telling actors when it is all right for them to succumb to their self-protective desires. We are asking whether they constrained those desires in order to avoid harming others to the extent we expect ordinary persons to do so. We are not asking whether they did the right thing: they did not. We are asking whether their having done wrong is excusable given their situation. For an excuse – whether one believes that excuses rest on assessments of the difficulty in choosing correctly or on the character reflected in the choices – a standard rather than rules seems appropriate. The situations actors will confront will be impossible to anticipate and to cabin in general rules.

On the other hand, although we have deemed our proposed defense an excuse, we are in fact equivocal regarding whether in some situations it is better seen as a personal justification. A personal justification would be a justification reflecting the moral permissibility of an actor’s giving more weight in the moral calculus to his and his family’s interests than those interests would be given from an impersonal perspective. In other words, the defense could reflect the fact that morally speaking, an actor may treat his life as more important than, say, the lives of five innocent aggressors, even though from society’s perspective their five lives morally outweigh his.

If our proposed defense is a personal justification rather than an excuse, then perhaps it could and should be “rulified.”69 If we can calibrate the extra moral weight that we can assign to our own interests in the moral calculus, then we could perhaps decide, say, that one may kill two but not more than two innocent aggressors to save one’s own life, and so forth.70

69In Chapter 8, when we take up the translation of our theory into criminal law doctrine, we return to the question of how much of our theory can be “rulified,” how that can be done, and whether it is desirable to do so.

70T he personal nature of the justification would still render it excuselike – that is, unjustified – insofar as third-party interveners and the targets of the actor’s risk imposition are

DEFEAT ER S OF CULPA BILI T Y

147

We here assume that the defense is, like duress, a pure excuse and thus avoid any speculation about how the defense might be reduced to rules. If that is the proper approach, then a “person of reasonable firmness” is the proper standard for assessing the actor’s self-protective crimes.

b. The Objectivity of the “Person of Reasonable Firmness” Standard: One of the perennial problems in criminal law is the question of whether the reasonable person should be altered to include the particular characteristics of the actor. Indeed, one of our primary complaints about liability for negligence is that there is no principled and rationally defensible way to determine with which features of the actor to endow the “reasonable person.” One objection to our proposal is that our “person of reasonable firmness” standard resurrects this problem. It does not. Let us explain why.

The “person of reasonable firmness” standard is wholly objective. It asks a normative question, that is, whether the actor has lived up to what society can reasonably expect of him.71 This standard does not change subject to the peculiarities of the actor.

Notice, however, that the characteristics of the actor will be part of the culpability calculation itself. For that calculation involves asking what risks did the actor perceive, and what reasons for the action was the actor aware of. Both of these questions are entirely subjective in their focus: the actor’s perceptions inform the risks of which she is aware, and her perceptions and motivations inform the reasons that she believes justify her behavior. It is only after both of these subjective features are placed on each side of the scale that an objective moral weighing is appropriate. Thus, the fact that a petite woman will feel threatened when a large man will not is captured by the risks and reasons that each individual actor will perceive and possess.

Moreover, although the question is whether the actor has shown the fortitude necessary to deem him nonculpable, there is some room for individualizing and thus subjectivizing in assessing the difficulty of his choice. Extreme phobias, philias, manias, addictions, and temptations

concerned. (Third parties might have to intervene on the side of the two innocent aggressors rather than the one personally justified defender.)

71Cf. Fletcher, supra note 4, at § 8.4.1 (discussing the German concept zumutbarkeit – what can fairly be expected of offenders); Peter Westen, “Individualizing the Reasonable Person in Criminal Law,” 2 Crim. L. & Phil. 137 (2008).

148	T HE CULPA BLE CHOICE

may render the consequentially preferred or deontologically demanded choice quite dysphoric for the actor. And because culpability is scalar rather than binary, such dysphoria could render the actor’s choice less culpable than it otherwise would have been or even completely nonculpable. We believe, however, that these impairments are better viewed as offering an independent excuse or mitigation for diminished rationality. We note, though, that it is possible that an actor will experience a “hard choice” because of his diminished rationality, leading to the conjunction of two different theories of excuse or mitigation. We leave this possibility open.

c.Probabilities, Retreat, and Proportionality: Our proposed defense of preemptive self-protection, with its single standard of a “person of reasonable firmness,” handles, in one fell swoop, the vexing problems of what probability of attack is necessary to trigger preemptive force, when the defender must retreat rather than use preemptive force, and whether the defender may use disproportionate force to prevent an attack. With respect to the first question, the actor is excused for using preemptive force whenever he estimates that the likelihood of attack on him is at a level at which a “person of reasonable firmness would use force self-protectively rather than wait for the probability of attack to increase. And, with respect to the interrelated doctrines of retreat and proportionality, viewing self-protection as an excuse rather than a justification helps support their presence in the law. The choice between killing an attacker and safely retreating is not a hard choice, nor is the choice between killing to protect one’s apples and losing one’s apples if proportionate force will not succeed. A “person of reasonable firmness” would safely retreat rather than kill and would lose his apples rather than kill. Or, put differently, it takes no extraordinary courage or resolve to forgo killing and suffer the minor losses of retreating or employing lesser force.72

d.Mistakes: Because the proposed defense of preemptive self-protection is an excuse, not a justification, it is immaterial whether the actor’s

72We leave to the side whether the retreat or proportional force or necessity requirement applies when the actor believes the attacker is a CA or CP.

DEFEAT ER S OF CULPA BILI T Y

149

assessments of the probability of attack, the likely consequences of attack, the time left to retreat, the safety of the retreat, the force necessary to resist, and so forth are correct or mistaken. The question is whether, given the actor’s beliefs on these matters, he acted as would a “person of reasonable firmness” in the situation as the actor assessed it. And if the actor’s beliefs were mistaken in a way such that a “person of reasonable firmness,” apprised of the mistake, would have acted differently from the actor and not committed the crime, the actor still may invoke the defense.

e. Third-Party Intervention: Because the proposed defense is an excuse or a personal justification, it is limited to the actor. Third parties who are not themselves threatened cannot stand in the actor’s shoes and invoke his excuse or personal justification to justify or excuse their conduct. Thus, if two innocent aggressors are attacking one innocent defender, the latter may be excused for using deadly force against the former. Third parties, however, if they may intervene at all, may not intervene with deadly force on the defender’s side of the struggle.73 Because the actor is only excused and not justified, and because the third parties are

73At least they may not do so as a matter of justification. If the actor were, say, a close family member, the third parties might themselves be able to invoke duress to excuse their aiding the defender rather than his attackers.

Jeff McMahan has argued that at least some of those whom we deem to be innocent aggressors can be justifiably harmed (in a social, not personal, sense of justifiability), even if they are more numerous than those they threaten. See McMahan, “The Basis of Moral Liability to Defensive Killing,” supra note 52, at 393–394. He gives as an example a driver whose car goes out of control nonculpably and threatens a pedestrian, who can save himself only by killing the driver. He contends that the pedestrian would be justified in the social sense were he to kill the driver. Presumably, as well, a third party could assist the pedestrian in killing the driver.

We disagree. McMahan assumes that the driver is threatening the pedestrian in a way that the pedestrian is not threatening the driver – that is, that this is a case of nonreciprocal risk imposition – and that this is what justifies the pedestrian and third-party aiders in killing the nonculpable driver. But this is incorrect. A pedestrian who is armed and prepared to kill nonculpable drivers who lose control is as much a threat to drivers as the drivers are to pedestrians. We think the driver’s lack of culpability is sufficient to make this a case of excuse or personal justification.

McMahan also claims that some victims of socially justified risk impositions are socially justified in resisting the socially justified risk imposers. Id. at 399. Again, we disagree. The one worker on the trolley siding is not justified in shooting the actor to prevent the actor from killing him but saving the five. Nor, contrary to what McMahan claims, may the victims of Strategic Bomber justifiably (as opposed to excusedly) attempt to shoot him down.

150	T HE CULPA BLE CHOICE

not in danger themselves, the third parties would be neither excused nor justified.

Moreover, a friend who assists the mother who is acting under duress is not a counterexample to this position. Rather, one reason why a friend might be entitled to assist the mother would be the friendship relationship. Thus, the friend may be personally justified in giving more weight to her friend’s interests, or the friend, too, may be excused because the emotional bonds of friendship bear on what a “person of reasonable firmness” would do.

5. A Possible Extension? Preemptive Collective Protection

and Preventive Detention

The excuse of preemptive self-protection, except for cases of situational duress, deals with commission of crimes on the basis of predictions of future human choices. Whether the actor is reacting to one who appears to be about to attack him or to one who has threatened him with a future attack, what gives rise to the excuse is the actor’s prediction that if he does not commit the crime, another will choose to aggress against him.

Actions that are preemptive in that way are troubling. We generally condemn preemptive restraints on liberty on the basis of predictions of future dangerous choices. There are exceptions, of course. We countenance not only self-defense but also restraining orders, peace bonds, gun restrictions, and restrictions on information that is dangerous in the wrong hands, and we do allow the preventive detention of those who, because of mental defects, will not be morally responsible for their predicted dangerous acts.

We draw the line, however, with preventive detention of those who are predicted to commit future crimes but who are fully responsible actors. We associate such preventive detention with totalitarian regimes, which paradigmatically act preemptively.

If, however, actors are excusable for acting preemptively in self-defense and duress situations, then is it possible that society as a whole could be excused for preventively detaining the sane but dangerous? The idea of a society’s being “excused” rather than justified may seem strange, but a society is nothing but those individuals who compose